Mode tracking method based on information entropy prejudging mode swapping

Abstract

Mode tracking based on correlation is a common method to achieve reasonable mode ordering in characteristic mode analysis (CMA). However, it is time-consuming to traverse the correlation calculation with all frequency points. For this reason, it is proposed to use information entropy in determining the frequency points of mode swapping in advance, which effectively improves the computational efficiency of CMA. Meanwhile, it is pointed out that for higher-order modes or electrically large structures, pseudo-degenerate modes are prone to appear, which results in mode swapping. In a large frequency range, the reactance property of a mode can be reversed, causing a change in the physical meaning of the mode. Therefore, CMA is best performed in the frequency range adapted to the structure size. Numerical analysis examples show the feasibility of this mode tracking method and confirm the rationality of mode performance analysis.